microPIECE - microRNA pipeline enhanced by CLIP experiments

All microRNAs are assumed to be post-transcriptional fine-regulators. With a length of around 21 nucleotides, they form a RNA-induced silencing complex (RISC) complex with a protein of the Argonaute family. This complex then binds to the messengerRNA untranslated regions and coding sequence regions and in general promotes degradation or translational inhibition. It is now important to know the microRNA-mRNA pairs in order to infer dysregulating effects on the organism. In order to assign a microRNA to a mRNA target, various tools with different technical approaches were developed. They are mostly based on the assumption that the first eight nucleotides of the microRNA (seed region) determine the binding region on the mRNA. Some approaches also include supporting bindings in the rear part of the microRNA, others take secondary structures of the mRNA or binding energies of the mRNA-miRNA complex into account. Nevertheless, they all suffer from the statistical problem that such short target regions, often occur simply by chance in transcript sequences. This results in a huge amount of false positive predictions. A target prediction of all 590 Tribolium castaneum mature microRNAs from miRBase.org v22 (Kozomara and Griffiths-Jones 2013) against all 18.534 protein coding cDNA sequences from Ensembl.org (Ensembl Genomes release 38 December 2017) (Kinsella et al. 2011) results in 2.948.255 possible microRNA-target interactions, predicted by the commonly used tool miranda (Betel et al. 2008) with standard parameters. To increase the credibility, wet lab validation methods like luciferase reporter assays are required. The disadvantage here is that this workflow is not applicable for high-throughput analysis, as it can only treat small subsets of sequence combinations. Another, more scalable method is cross-linking immunoprecipitation-high-throughput sequencing (CLIPseq). Here, binding regions of the RISC show a specific signal in the sequencing reads that can be used to shrink the search space of miRNA target predictions, when mapping them to the transcriptome. The limitation here is the difficult technical treatment in the laboratory. This is the reason why there are only a few datasets available for human, mouse, worm and mosquito. It would now be useful, if we could simply transfer the information of a binding region, already identified by CLIP-seq, to another species. This is what our microRNA pipeline enhanced by CLIP experiments microPIECE is about.